Active Textile Antennas as a Platform

for More Energy-Ecient and Reliable

Wireless Links in Healthcare
Hendrik Rogier1,2, Frederick Declercq1,2 , Patrick Van Torre1 ,
and Luigi Vallozzi1
1

Ghent University, Dept. of Information Technology

Sint-Pietersnieuwstraat 41, B-9000 Ghent, Belgium
Hendrik.Rogier@intec.UGent.be
2
IMEC, INTEC Division, Ghent, Belgium

Abstract. New wireless wearable monitoring systems worn by patients

and caregivers require a high degree of reliability and autonomy. We
show that active textile antenna systems may serve as robust platforms
to deploy such wireless links, in the meanwhile being comfortable and
invisible to the wearer, thanks to recent developments in the design process combined with dedicated signal processing techniques. The key idea
is to exploit the large amount of real estate available in patients and
caregivers garments to deploy multiple textile antennas each with a size
large enough to make them ecient radiators when deployed on the body.
The antenna area is then reused by positioning active electronics directly
underneath and energy harvesters directly on top of the antenna patch,
ensuring the autonomy of the module. Combining dierent antenna signals by means of low-power multi-antenna processing techniques then
ensures good signal quality at low transmit power in all situations.
Keywords: Smart Fabrics and Interactive Textiles, wearable electronics, active textile antennas, MIMO signal processing.

Introduction

The explosive developments in the eld of smart fabrics and interactive textiles
(SFIT) have opened a wide range of exciting possibilities in the eld of healthcare. Remote monitoring of patients may be provided by wearable systems that
are unobtrusively integrated into patient garments, in the meanwhile continuously monitoring life signs, activities and environmental conditions, and relaying
these data wirelessly to a remote location for supervision by a caregiver. Such
intelligent garments may be deployed both in the hospital and in the home environment of the wearer. Also caregivers may be equipped with wearable systems,
either to collect and transmit their own parameters during interventions in hazardous situations, or as a personal wearable computer that receives, interprets
and displays patient data to help a caregiver in making decisions.
B. Godara and K.S. Nikita (Eds.): MobiHealth 2012, LNICST 61, pp. 318325, 2013.
c Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2013


Active Textile Antennas for Healthcare

319

As SFIT systems are to be deployed in critical applications such as patient

monitoring and rescue missions, their reliability and autonomy are two key concerns of the designers. To ensure sucient autonomy without the use of heavy
batteries, we need to satisfy two requirements: First, the wireless communication
module, and in particular the transceiver module responsible for setting up the
wireless link is typically one of the largest energy consumers of the wearable
system. Therefore, we need to design a highly energy-ecient communication
system, in which textile antennas are critical components. Indeed, an ecient
wearable antenna providing o-body communication with high gain and large
radiation eciency typically consumes quite a lot of space, as a large ground
plane is required to avoid absorption of antenna radiation by the human body.
Yet, this space is widely available in garments. However, the large exible textile
antenna [6, 4, 5, 3] will typically be subject to bending, wrinkling and crumpling
as the wearer moves around. This should be taken into account during the design phase of these antennas. Second, to ensure sucient autonomy to the SFIT
system, energy harvesters should be added that make use of the large number of
energy sources available in the neighbourhood of the body to scavenge energy.
In this contribution, we will review and evaluate a number of techniques that
may be implemented on textile antennas to increase the energy-eciency and
autonomy of SFIT systems. The key idea is to increase performance and functionality by reusing surface area for multiple purposes. This is done at two levels:
First, the garment, oering protection and comfort to the wearer, is reused as a
platform for a wearable multi-antenna system. Second, the textile antenna itself,
being the largest component of each wearable module, is reused as a platform
for active electronic circuits implementing sensing and communication, and for
energy-scavengers with energy-management circuits.
In Section 2 we show how active electronic circuits may be directly integrated underneath the wearable antenna, resulting in a compact communication
module. Moreover, by adopting a full-wave/circuit co-design and co-optimizing
strategy, optimal active antenna performance can be obtained, aiming either for
optimal noise or impedance matching. Next, we show in Section 3 how one can
integrate a set of solar cells on top of the antenna patch, without disturbing the
radiation characteristics of the textile antennas. In Section 4 we show that by
combining multiple wearable antennas, integrating them in the front and back
sections of a garment and relying on diversity techniques at both ends of the
link, the reliability of wireless communication links can be improved tremendously. Diversity techniques can easily be implemented at the receiver side, but
the use of space-time codes can be quite energy-consuming as codes must be
generated at the transmitting side, whereas accurate channel estimation is required at the receiver. Section 5 presents some energy-ecient alternatives, such
as passive beamforming and channel state tracking. It is shown that these techniques provide an almost equal performance while relying on much simpler and
less energy-consuming signal processing algorithms. Finally, we formulate some
conclusions in Section 6.

320

H. Rogier et al.





 

 
 
 

Fig. 1. Active wearable receive antenna operating in 2.45GHz ISM band

Full-Wave/Circuit Co-design of Active Wearable

Antennas

Although the development of wearable electronics systems has boomed over the
last decade, most of the research focused on converting conventional rigid electronic circuits into exible wearable components that are compatible with integration into fabric. The complete wearable electronic system is then constructed
by interconnecting the set of wearable components needed to implement the desired functionality. This approach, however, results in a fragile overall system,
with many weak links that easily break. Moreover, as all components are designed separately, optimal performance of the overall system is not guaranteed.
Therefore, we have adopted a new design strategy [2] where we aim to integrate
as many components as possible on the textile antenna(s) of the wearable system, in the meanwhile reusing the large area that we need to ensure high gain
and good radiation eciency, and opening new opportunities to co-optimize the
performance of the antenna together with the active circuits. At the same time,
improving overall system performance also increases the autonomy of the wireless
system and reduces the risk of interference with other systems. Both problems
are highly relevant in healthcare applications.

Active Textile Antennas for Healthcare



321



 


 
 
 















 







 


 














Fig. 2. Aperture-coupled shorted wearable solar patch antenna for 902928 MHz UHF
band

In Fig. 1 we present an active wearable antenna, for which the full-wave/circuit

co-design approach was applied to integrate a low-noise amplier (LNA) directly
underneath a textile patch antenna. As the antenna feed is directly connected
to the LNA input by means of a short via, RF connection losses are reduced.
Moreover, careful co-optimization of the patch dimensions, the position of the
antenna feed point and the elements composing the LNA circuit results in optimal noise performance, high gain and 50 matching at the output of the LNA,
leading to a highly sensitive receive antenna that operates in the 2.45GHz ISM
band.

Integration of a Solar Cell on a Wearable Antenna

Ensuring acceptable autonomy is of major concern for wearable systems, since

batteries rapidly increase the weight and may reduce the exibility. An exciting
new area of the research focusses on harvesting energy from the body and its
environment to power the system. In particular, solar energy and kinetic energy
originating from body movement are important sources the wearable system may
rely upon. In addition to placing active electronics below the antenna patch, we
now also integrate energy harvesters directly on top of the wearable antenna
module [1]. In Fig. 3 we present an aperture-coupled shorted wearable solar
patch antenna for communication in the 902928 MHz UHF band.

322

H. Rogier et al.

MIMO Techniques for Wearable Systems

In indoor environments, wireless communication systems suer from multipath

fading, which signicantly reduces the reliability of the data link. In case of
wearable SFIT units, their deployment in the direct neighborhood of the human
body results in important additional shadowing eects. To overcome shadowing
and fading, typically higher transmit powers are required to achieve acceptable
Bit Error Rates (BERs), which seriously reduces the autonomy of the wearable system. However, by making use of a garment as a platform for the integration of multiple wearable transmit/receive units and by relying on MIMO
diversity schemes [8], highly reliable links may be implemented using low transmit power. Fig. 3 demonstrates the improvement in terms of BER for dierent
MIMO schemes that were implemented using a wearable system consisting of
two dual-polarized textile antennas integrated in a professional garment. The
link quality was studied for a person walking around in an indoor oce environment in non-line-of-sight conditions. It is, however, important to approach BER
results based on measured instantaneous signal levels like the ones obtained in
Fig. 3 with care. In reality, the BER for the real data demodulation suers from
imperfect channel estimation in case of low signal levels. In case of erroneous
channel estimation, the space-time codes will not oer the full diversity gain
expected from theory. A transmission of 100 frames with channel estimation by
300 BPSK pilot symbols at an average SNR of 10dB resulted in a BER at the
receiver of 1.7e-2 for a 4 1, 8e-3 for a 4 2 and 1.7e-3 for a 4 4 link. Clearly,
the measured improvements in wireless link quality are found to be much lower
than expected from theory. It was found that, by just increasing the average
SNR from 10dB to 12dB, for the set of 100 transmitted frames the number of
bit errors dropped to zero for the 4 4 communication link.

Energy-Ecient Multi-antenna Processing Techniques

Although space-time coding (STC) via multiple antenna oers large potential
benets in terms of link quality, it comes at a cost in terms of required processing power to manipulate the codes and to perform accurate channel estimates.
Therefore, we have investigated an alternative approach where simple static
beamforming is performed using a four-element textile antenna array that transmits a beam conned in the vertical direction [7], resulting in a larger received
signal level and reducing the number of received multipath components. This
in turn decreases the uctuations in received signal level. In Fig. 4 we compare
space-time coding and static beam forming in non-line-of-sight conditions in an
indoor environment. Moreover, BER versus SNR curves are presented for two
receiver congurations: a single-antenna receiver without diversity versus a fourantenna receiver implementing maximum ratio combining (MRC). The curves
were normalized to include TX array gain, which corresponds to ensuring an
equal total transmitted energy per information bit Eb,tr at the transmitter. In absence of receiver diversity, the space-time code clearly outperforms beamforming

Active Textile Antennas for Healthcare

323

0
10

BER

2
10

4
10

6
10
10

1x1 No diversity
2x2 Alamouti Pol.
2x2 Alamouti F/B
4x4 OSTBC
Rayleigh; 1x1
Rayleigh; 2x2
Rayleigh; 4x4
5

0
5
E /N per input antenna [dB]
b 0

10

Fig. 3. Bit Error Rate versus Signal to Noise Ratio for various orders of MIMO systems implementing front/back (F/B) and/or polarization (Pol.) diversity by means of
Orthogonal Space-Time Block Codes (OSTBCs). The curves derived from measured
instantaneous signal levels for a person walking in an indoor environment are compared
to the theoretically achievable gains in case of ideal Rayleigh fading.

10

BER

10

10

Rayleigh
10

STC (no div.)

Beam (no div.)
STC (MRC)

10

4
5

Beam (MRC)
0

5
E /N
b 0

10

Fig. 4. Bit Error Rate versus Signal to Noise Ratio: comparison between Space-Time
Coding (STC) and beam forming (Beam) with 4th order receive diversity (MRC combining) and in absence of diversity

324

H. Rogier et al.

10

BER

10

10

SISO (16)
Rayleigh
Data CSI
Pilot CSI

10

6
10

Perfect CSI
5

0
5
10
Avg. E /N per input antenna
b 0

15

20

Fig. 5. Bit Error Rate versus Signal to Noise Ratio for the 4 4 MIMO channel:
comparison between perfect channel knowledge, estimation based on pilot symbols and
data driven channel tracking

as for the higher Eb /N0 values, the transmit diversity gain rapidly exceeds the
beamforming power gain. In case of receiver diversity, however, the beamforming BER is lower than the BER for space-time coding over the full range we
considered.
If we prefer the benets of additional diversity gain oered by the space-time
codes, we need to implement techniques that alleviate the additional processing
cost. One approach consists of reducing the overhead in terms of pilot symbols
that are needed to make accurate channel estimates at the receiver. This is done
by implementing data driven channel tracking [9], where we use the detected
data symbols as substitutes for the pilot symbols to keep track in time of the
correct channel state. Fig. 5 compares the performance in terms of BER of a
4x4 MIMO system that has perfect channel knowledge at the receiver with an
implementation that makes abundant use of pilot symbols for channel estimation
(300 pilot symbols transmitted in separate time-slots for each transmit channel
in addition to 528 symbols of space-time coded data per frame) and a link for
which data driven channel tracking is implemented at the receiver. We observe
that, once the BER drops below 2e-4, the STBC-MIMO system with channel
tracking oers a performance similar to that of the MIMO system using large
amounts of pilot symbols for channel estimation.

Active Textile Antennas for Healthcare

325

Conclusion

It was shown that garments of patients and caregivers may be used as platforms
to deploy wearable multi-antenna systems that sense the wearer and his/her
surroundings. Direct integration of active electronics and energy scavengers onto
the active textile antennas lead to compact and energy-ecient modules that are
light-weight and comfortable to wear.

References
[1] Declercq, F., Georgiadis, A., Rogier, H.: Wearable Aperture-Coupled Shorted SolarPatch Antenna for Remote Tracking and Monitoring Applications. In: Fifth European Conference on Antennas and Propagation - EuCAP 2011, Rome, Italy, pp.
29922996 (April 2011)
[2] Declercq, F., Rogier, H.: Active Integrated Wearable Textile Antenna With Optimized Noise Characteristics. IEEE Trans. on Antennas and Propagation 58(9),
30503054 (2010)
[3] Hertleer, C., Rogier, H., Vallozzi, L., Van Langenhove, L.: A textile antenna for
o-body communication integrated into protective clothing for reghters. IEEE
Trans. on Antennas and Propagation 57(4), 919925 (2009)
[4] Locher, I., Klemm, M., Kirstein, T., Troster, G.: Design and characterization of
purely textile patch antennas. IEEE Transactions on Advanced Packaging 29(4),
777788 (2006)
[5] Salonen, P., Rahmat-Samii, Y.: Textile antennas: Eects of antenna bending on input matching and impedance bandwidth. IEEE Aerospace and Electronic Systems
Magazine 22(12), 1822 (2007)
[6] Tronquo, A., Rogier, H., Hertleer, C., Van Langenhove, L.: Robust planar textile
antenna for wireless body LANs operating in 2. 45 GHz ISM band. IEE Electronic
Letters 42(3), 142146 (2006)
[7] Van Torre, P., Scarpello, M.L., Vallozzi, L., Rogier, H., Moeneclaey, M.,
Vande Ginste, D., Verhaevert, J.: Indoor O-Body Wireless Communication: Static
Beamforming versus Space-Time Coding. Intern. Journal of Antennas and Propagation, Article ID 413683 (2012)
[8] Van Torre, P., Vallozzi, L., Hertleer, C., Rogier, H., Moeneclaey, M., Verhaevert,
J.: Indoor O-Body Wireless MIMO Communication With Dual Polarized Textile
Antennas. IEEE Trans. on Antennas and Propagation 59(2), 631642 (2011)
[9] Van Torre, P., Vallozzi, L., Rogier, H., Moeneclaey, M., Verhaevert, J.: Channel
Characterization and Robust Tracking for Diversity Reception over Time-Variant
O-Body Wireless Communication Channels. EURASIP Journ. on Advances in
Signal Processing, Article ID 978085 (2010)